*** In the eyes of the beholder? Understanding policymakers’ perceptions of corruption

*** Cristian Márquez Romo & Xavier Romero-Vidal

*The following is a replication file for the tables and models found in the article. 
*Please contact Cristian Márquez (cmarquez@usal.es)
*for additional materials related to the study (e.g., data
*visualization, coding of variables or appendix materials)

*Figure 1. Average perceived corruption level vs. average political corruption
collapse PRO112 v2x_corres, by(country)
twoway (scatter PRO112 v2x_corres, mlabel(country)) || lfit PRO112 v2x_corres, scheme(s1mono)

*Table 1. Individual and Country-Level Drivers of Corruption Perceptions Among Legislators
xtmixed PRO112 c.year i.government c.v2x_corr c.GDPgrowth c.unemployment i.reelected c.ID1 i.gender c.age i.education || country: , mle var
	est sto m1

xtmixed PRO112 c.year c.v2x_corr c.GDPgrowth##i.government c.unemployment i.reelected c.ID1 i.gender c.age i.education || country:government, mle var
	est sto m2
	
xtmixed PRO112 c.year c.v2x_corr c.GDPgrowth c.unemployment##i.government i.reelected c.ID1 i.gender c.age i.education || country:government, mle var
	est sto m3

esttab m1 m2 m3, se(2) b(2) star(* 0.1 ** 0.05 *** 0.01) label
	
	
*Figure 2. Predicted average marginal effects of government-opposition status on corruption perceptions, conditional on the economic cycle (GDP growth)
xtmixed PRO112 c.year c.v2x_corr c.GDPgrowth##i.government c.unemployment i.reelected c.ID1 i.gender c.age i.education || country:government, mle var

margins, at(GDPgrowth = (-2(1)10) government = (0 1))
marginsplot, recast(line) plot1opts(lcolor(gs8)) ciopt(color(black%20)) recastci(rarea) xtitle("GDP growth (annual %)") scheme(s1mono)

*Figure 3. Predicted average marginal effects of government-opposition status on corruption perceptions, conditional on the economic cycle (Unemployment rate)
xtmixed PRO112 c.year c.v2x_corr c.GDPgrowth c.unemployment##i.government i.reelected c.ID1 i.gender c.age i.education || country:government, mle var

margins, at(unemployment = (2(1)10) government = (0 1))
marginsplot, recast(line) plot1opts(lcolor(gs8)) ciopt(color(black%20)) recastci(rarea) xtitle("Unemployment (annual %)") scheme(s1mono)

*Appendix A:

*Figure A1. Average perceptions of corruption among legislators
collapse PRO112 v2x_corres, by(country year)

twoway (connected PRO112 year), scheme(white_tableau) by(country)

*Appendix B: robustness checks

*Table A3. Individual and Country-Level Drivers of Corruption Perceptions Among Legislators (fixed-effects models)
reg PRO112 c.year i.government c.v2x_corr c.GDPgrowth c.unemployment c.inflation i.reelected c.ID1 i.gender c.age i.education ib10.country, robust
	est sto am1a
	
reg PRO112 c.year c.v2x_corr c.GDPgrowth##i.government c.unemployment c.inflation i.reelected c.ID1 i.gender c.age i.education ib10.country, robust
	est sto am2a
	
reg PRO112 c.year c.v2x_corr c.unemployment##i.government c.GDPgrowth c.inflation i.reelected c.ID1 i.gender c.age i.education ib10.country, robust
	est sto am3a

reg PRO112 c.year c.v2x_corr c.inflation##i.government c.GDPgrowth c.unemployment i.reelected c.ID1 i.gender c.age i.education ib10.country, robust
	est sto am4a

esttab am1a am2a am3a am4a, se(2) b(2) star(* 0.1 ** 0.05 *** 0.01) label
	
*Table A4. Individual and Country-Level Drivers of Corruption Perceptions Among Legislators (multilevel models/w inflation)
xtmixed PRO112 c.year i.government c.v2x_corr c.GDPgrowth c.unemployment c.inflation i.reelected c.ID1 i.gender c.age i.education || country: , mle var
	est sto am1b

xtmixed PRO112 c.year c.v2x_corr c.GDPgrowth##i.government c.unemployment c.inflation i.reelected c.ID1 i.gender c.age i.education || country:government, mle var
	est sto am2b

xtmixed PRO112 c.year c.v2x_corr c.unemployment##i.government c.GDPgrowth c.inflation i.reelected c.ID1 i.gender c.age i.education || country:government, mle var
	est sto am3b
	
xtmixed PRO112 c.year c.v2x_corr c.inflation##i.government c.GDPgrowth c.unemployment i.reelected c.ID1 i.gender c.age i.education || country:government, mle var
	est sto am4b

esttab am1b am2b am3b am4b, se(2) b(2) star(* 0.1 ** 0.05 *** 0.01) label

*Figure A2. Predicted average marginal effects of government-opposition status on corruption perceptions, conditional on the economic cycle (Inflation)
margins, at(inflation = (2(1)10) government = (0 1))
marginsplot, recast(line) plot1opts(lcolor(gs8)) ciopt(color(black%20)) recastci(rarea) xtitle("Inflation (annual %)") scheme(s1mono)
